1 of 1 people found this helpful
How it works:
APs are polled from Controller. In case the AP is monitored with NPM, but is not in the results from the last poll, AP is removed from NPM. When it appears later it's added again as a new row in Wireless_AccessPoints table.
So when you reboot controller it looks like it takes some time then it starts to send APs in polled results again. So we remove all APs and later add them again.
What you can do here:
1. each time you will distribute a software upgrade to our controllers you should Unmanage Controllers in NPM and Remanage after controller is rebooted and is up and running
2. edit config file SolarWinds.Wireless.Collector.dll in C:\Program Files (x86)\SolarWinds\Orion\Wireless and change the value for RemoveDisappearedAps to False. Restart services.
Now APs will be only marked as Disappeared and Up when they come back later.
Thanks for the reply --
1. We have over 9000 WAPs and 30 controllers; unmanaging doesn't work as that stops statistics collection.
2. I can apply this option temporarily but how to I now remove the APs?
The AP hasn't disappeared, it's just seamlessly migrated to its backup controller -- the users probably won't even have noticed their traffic glitched.
Now I seem to have started getting the same AP in the table twice:
SELECT name,ipaddress,count(*) FROM [dbo].[Wireless_AccessPoints] group by name,ipaddress having count(*)>1
HMC.6.1 172.20.x.y 2 HMC.6.2 172.20.x.z 2
It works exactly how it was designed.
There are only two options:
1. remove disappeared APs from database (disappeared from NPM point of view mean-the AP is not in results polled from controller)
2. don't remove disappeared APs
There is no way how to remove the APs on UI. You can change the key in config file back to True and they will be removed with next poll.
Anyway there is always a way how to make the feature better, so if you have some ideas please create a feature request.